As the U.S. population ages and health care costs continue to rise, policies aimed to control costs have become a major focus of the reform effort. Provider payment reform is at the core of these policies. The overall aim of this research is to understand the implications of provider payment reform on health care costs and physician behavior. The proposed research is largely empirical, but there is an economic theory component. This project will center on an empirical evaluation of a large payment reform policy in a private plan environment. Claims and enrollment data will be used to study the effects of changing a physician group's reimbursement towards prospective payment on costs, utilization, and physician referral behavior. The project will also develop two simple economic models, a normative model of physician optimal referral and a positive model of physician specialization, to complement the empirical work.